Visiontek Radeon HD 4850
Posted 08/14/08 at 02:00:00 PM | by Will Smith
Last month, we spent a ton of time talking about the efficiency and overall pixel-pushing prowess of ATI’s new GPU, so we won’t waste much ink on the subject here. Suffice it to say, the 4850 delivers enough power to drive your sweet, new 22-inch monitor at its native resolution.
Visiontek’s Radeon HD 4850 delivers entry-level DirectX 10 performance at a compelling price.
The card’s silicon is equivalent to that of previous-gen high-end cards. It’s equipped with 512MB of GDDR3 memory running at 993MHz. Unlike the Radeon HD 4870 boards (which cost $100 more), the 4850 doesn’t sport GDDR5 (GDDR5 transfers twice as much data per clock cycle as GDDR3). The upshot? The HD 4850 has the slowest memory interface of any card in the current generation, and benchmarks show that—especially at high AA/anisotropic filtering levels.
The HD 4850 does sport the same GPU as the 4870, but it’s clocked down to a modest 625MHz. Unlike the lesser Nvidia parts, which feature fewer stream processors, the 4850 includes a full complement of 800 stream processors paired with 40 texture units, just like the 4870. This means the HD 4850 is at its best in shader-heavy benchmarks such as Crysis.
At the $200 price point, this card’s main competition is the old GeForce 8800 GT/9600 GT line of parts, against which it compares favorably. In benchmarks that are limited by shader performance, the 4850 absolutely slaughters the older GPUs. In memory-bandwidth-limited benchmarks, the older GPUs close the gap. While the benchmarks we list are primarily geared toward high-resolution screens, we also run some lower-resolution tests—Crysis on Very High chalked up a respectable 15.3 fps, on High it averaged 28 fps. Our image-quality tests didn’t show any anomalies, and high-def video playback was flawless.
For anyone riding an old DirectX 9-era GPU, the HD 4850 is your ticket to full DirectX 10 capability—and a more than capable upgrade from your old card. For folks who already own a DirectX 10 card, there’s really nothing to see here.
VisionTek Radeon HD 4850
Dr. Horrible
Cheap-and-easy DirectX 10. Single-slot. Supports HDMI, HDCP, and accelerated Blu-ray playback.
Bad Horse
Not that much faster than an 8800 GT. Conks out at high resolutions, AA/anisotropic filtering levels.
Benchmarks
| Radeon HD 4850 | EVGA GeForce 8800 GT | |
|---|---|---|
| 3DMark Vantage Game 1 (fps) | 8.62 | 6.0 |
| 3DMark Vantage Game 2 (fps) | 7.34 | 5.69 |
| Crysis (fps) | 11.9 | 4.4 |
| World in Conflict (fps) | 11 | 16 |
| 3DMark06 Game 1 (fps) | 29.3 | 25.2 |
| 3DMark06 Game 2 (fps) | 23.9 | 23.3 |
Best scores are bolded. Benchmarks are run on our videocard test bed, which consists of a QX9800 CPU, an Intel X48-chipset motherboard, and 4GB of DDR3 memory. All tests run at 1920x1200 with 4x AA and 16x anisotropic filtering enabled. Crysis is run at Very High settings.
Comments
Print
Email
Septemper Print Issue?
Submitted by guest001 on Sat, 2008-10-04 10:12
This review confused me greatly, considering it gave me mixed messages from the september issues. However, there is one very strange thing I noticed: the fps in World in Conflict dropped 20 frames running at the same resolution and AA. Is this an issue with the print copy or the web copy, because almost all of the other benchmarks are extremely similar.
Submitted by CWarren410 on Sat, 2008-08-16 03:50
Does anyone know which applications are capable of using the hardware decoding on this card? I bought another Radeon card over a year ago that promised hardware support only to find that no one supported it. Thanks, Charles Warren III
Submitted by karnak on Fri, 2008-08-15 06:39
Will Smith should face some disciplinary action from putting out this little piece of writing. Incorrect testing methodology, misuse of words, downright lying! (QX9800 CPU??) We're all human, but I expect more from my Maximum PC!
Time to go check all of Will's older articles...
Submitted by karnak on Fri, 2008-08-15 06:41
So I guess the EDItOR IN CHIEF of maximum PC must be an NVIDIA fanboi or something to write this piece of crap.
Submitted by yendor on Fri, 2008-08-22 11:37
Yea, I am getting the feeling Maximum PC shows a bias towards nvidia. 7?
I have never heard maxPc write things like "Not that much faster than an 8800 GT" as a con. They always praise products for being just a little bit better.
WHAT?
Submitted by Strongbad536 on Thu, 2008-08-14 20:31
So its 7, but yet its on the best of the best? Will, a sub $200 card is not meant for 24" or even 30" resolutions. You should have done them at 1440x900 or 1680x1050, and done many more games. And to say its only modestly faster than the 8800GT, but its 2.5 times faster in the most demanding game out there.
Submitted by bholstege on Thu, 2008-08-14 13:30
This review was absolutley horrible, for the following reasons:
1) You only used 2 real games. I could care less what it gets in 3dMark. I don't play 3dmark.
2) It currently costs around $150, which is an insane value.
3) Of those 2 games, it absolutley destroyed the 8800GT in Crysis. As for WIC, it is often heavily CPU limited. I can't tell if that is the case, because you tested it on a "qx9800" which DOESN'T EXIST.
4) In actual games, this card has been shown to outpreform the 8800gt by 20%, while using less power, at an only slightly highe retail rpice.
5) a $150 card is not supposed to be used on a $500 monitor with AA and AF. Test it at 1440x900 and 1680x1050
And please, do reviews with actual games. This one sucked.
To their credit, they have
Submitted by jwalch.hawk on Thu, 2008-08-14 18:18
To their credit, they have been using benchmarks in this manner (even if the details, exact tests, etc. change periodically) forever. They have always used a combination of "synthetic" and "real world" tests. What I'm saying is that your first issue with this review aren't relegated to just this review, but is an issue that you have with their review process for video cards in general. I'm not saying that means you're wrong; I'm just clarifying that they've always reviewed this way. (#1 and #4)
Also, the QX9800 doesn't exist for us mere mortals yet, but I'm pretty sure it's one of those soon-to-be-released by Intel parts (boy, I should have a link to back this up). I think MaxPC was actually trying to remove the CPU from the equation by using the fastest part they could get their grubby little hands on. (#3)
Your other points (#2 and #5) all pretty much pertain to value-based arguments (I'm lumping the choice of monitor in here with this for a reason). MaxPC has historically not put very much of an emphasis on that (though that's not to say they ignore it entirely; just not a point of emphasis). The premise is that things should be compatible with the best things money can buy. I'm sure they knew full well going into the review that a $150 card was likely going to struggle on that monitor with AA and AF at high res. The counterargument here is that it doesn't matter that it you just can't make a $150 card that can do that - If you want the best, you have to expect that it can handle the craziest crap you'd want to do with it. The idea is that the reviews aren't relative, but absolute in terms of the scale. A 9 is a better overall product than 7. They leave it up to you to decide if saving $200 and buying the 7 is the better choice. Myself, I almost always would go with the 7. 7 is a pretty respectable score in terms of MaxPC reviews, and if I can find a value-priced product in that area, I'm gonna jump all over it. That doesn't necessarily mean that the 9 card can't do things that maybe the 7 card wasn't designed to do well (or at all).
Submitted by bholstege on Thu, 2008-08-14 18:30
I'm not sure that last bit is true. Didn't the 8800GTX get a higher score than the 8800 Ultra?
I really think they need to look at their testing methodology, and move to a more real world review set.
typo?
Submitted by band geek on Thu, 2008-08-14 12:27
"The HD 4850 has the slowest memory interface of any card in the current generation,"
Did you mean slowest? I thought it had some of the fastest-clocked memory.
The 4870 has the
Submitted by n0t_a_n000b on Tue, 2008-08-19 06:12
The 4870 has the superclocked memory, which is GDDR5, but the 4850 just has GDDR3, like all Nvidia overpriced crap.
N0t a n00b
Submitted by BaggerX on Thu, 2008-08-14 12:45
"The upshot? The HD 4850 has the slowest memory interface of any card in
the current generation, and benchmarks show that—especially at high
AA/anisotropic filtering levels."I'm not getting it either. How would being the slowest, and the fact that it "Conks out at high resolutions, AA/anisotropic filtering levels" as stated in the verdict section, be an upshot?
Submitted by n0t_a_n000b on Tue, 2008-08-19 06:09
Yeah, it conks out on high AA, but if you play games like that, buy the 4870 cause it has GDDR5, not GDDR3.
N0t a n00b
Submitted by jwalch.hawk on Thu, 2008-08-14 18:24
Yeah, by the pure dictionary definition, "upshot" just means more or less means "end result" of "consequence"... The connotation gets confusing, because I've heard some folks use it in a positive context (this is good) and some in negative (this is bad). Sometimes it can be hard to tell which was intended, though I'm almost certain that it was being used in a "this is bad" context here.
So I vote we ban the word "upshot." Who's with me? Anyone, anyone? Bueller...
Maybe this is why no one's ever told me I should be in charge of what words are acceptable in English and which ones aren't...
I think he meant it like,
Submitted by StormEffect on Thu, 2008-08-14 12:58
I think he meant it like, "What's the takeaway from all this?"
So he said, "[What's] the upshot?" as in, what does the GDDR3 memory mean in performance.
Yeah, probably could've used another phrase.
No, he's right
Submitted by Skiplives on Thu, 2008-08-14 16:11
The GDDR3 in the 4850 is slow compared to the memory clocks on nVidia cards or the 4870's GDDR5. The processor has tons of very good and fairly fast shader rendering units. So on a shader heavy game, like Conan or Crysis it cranks, especially compared to an older card like the 8800GT. If the game required a lot of memory to perform, the 4850 in spec form doesn't outshine the older cards. With twice the RAM maybe we see less of an issue, but that wasn't the card they reviewed.
Like the review says, if you are buying new or upgrading from a DX9 card, it is a great buy. If you own a 8800GT there is no reason to buy this new card, the playable difference is just not worth it. You are better off saving another $100 for the 4870.
________________________
That's my story, and I'm sticking to it.
RESOURCE CENTER
THIS MONTH's ISSUE 
FEATURE Windows Tips: Find out what works and what doesn't as we test the most commonly prescribed Windows tipsHOW TO Customize and streamline your Windows desktop Core i7 Check out Intel's next-gen chip, up close and personal The Reactor We preview the first production-ready oil-immersed PC
Technology News
Computer Cooling Fans
Computer Cases
PC Game Controllers
PC Games
Computer Hardware
Headphones
MP3 Players
Stream Video
Computer Mouse
Monitors
Motherboards
NAS Storage
Networking
Laptop Computers
DVD Burner
Digital Cameras
Portable Storage
Computer Accessories
Smartphone
Antivirus Software
Sound Cards
Speakers
Computer Systems
Thumb Drives
Video Cameras
Video Card Reviews
Water Cooling
Gadgets
- Keyboards
© 2008 Future US, Inc. All Rights Reserved.